jonathan@researchcomputingteams.org
In this interview, James Hetherington, Jonathan Cooper, Donna Swann, and Chris Langridge from UCL ARC talk about the work they’ve put in over the past several years to improve hiring, retention, professional advancement, and job role clarity on their team.
They went about it quite methodically, taking small steps which at each point solved some problem for them, and laid the foundation for the next:
I really enjoyed the conversation, and I hope you do too!
[00:01:07] Jonathan Dursi: Hi, everyone. I’m here speaking with the University College London Advanced Research Computing team. In issue 148 we had shared the story about the new "always open" vacancies at UCL ARC. And even more excitingly, it turns out that’s just the culmination of a lot of work that’s gone into defining professions within research computing and data and defining career ladders. So I’m here to talk with the team today about how this came to be and the problems that it solves. So let’s start with introductions. James would you like to introduce yourself?
[00:01:43] James Hetherington: I am the director of the UCL Advanced Research Computing center. And I founded the Research Software group at UCL about 10 years ago. But a few years ago, I went off to the Alan Turing Institute where I built a research engineering department at the Alan Turing Institute. I came back to UCL a couple of years ago to found the new Advanced Research Computing department at UCL, which is a hybrid professional services/ academic entity, which is in itself very unusual. And the old Research Software group which I founded, Jonathan continued, is now part of this new department. We’ve been working for many years to build the professional structures for the kinds of people that contribute to using data and computers and software to do science and scholarship and create a great home for them inside a university. And I think with. With UCL ARC we’re on the right direction towards cracking it.
[00:02:42] Jonathan Dursi: Thank you. And Donna, could you introduce yourself?
[00:02:45] Donna Swann: Hi, I’m Donna Swan. I’m head of the operations and strategy group. Also cheif of staff for ARC. So I’m on the senior leadership team and I head up here the operations and strategy sort of thing. So making sure the department is run from a, both a service delivery point of view, but also research and teaching.
Under me, I have around -is that 20 now in our group. Chris being our departmental manager. And under Chris, there’s the finance and administration roles. And then we have quite large delivery group now with a portfolio delivery manager with agile project managers sitting underneath and delivery support. So yeah, we went from two, I think, at the start to about 20 or so of us now. We have a community lead, research enterprise manager and an education lead as well.
[00:03:25] Jonathan Dursi: That’s terrific. And Jonathan, how about you?
[00:03:33] Jonathan Cooper: Hi, I’m Jonathan Cooper. I’m the head of Research Software Engineering within ARC. And I also lead our collaborations team, which is the part of ARC that runs projects with others, essentially collaborative projects, primarily with UCL researchers. And that’s pretty much where the Research Software group started, working on UCL research. But now we do collaborations beyond UCL as well. And indeed as ARC have the explicit permission to lead research in our own into better ways of using technology to support the research endeavor and improving how we ourselves work. So we’re I guess in some ways the research arm of ARC.
But actually we try to see everything that ARC does as being capable of producing research outputs that can be disseminated. It’s not just when we’re explicitly working on research grants, although that is a lot of what we do.
I came to UCL six years ago when I got tired of trying to persuade another institution that shall remain nameless that RSE was a good thing and that they should pay for it centrally.
[00:04:38] James Hetherington: An extremely old and distinguished UK university.
[00:04:41] Jonathan Cooper: Yes. And joined UCL. And yeah. Pretty much enjoyed my time here since. And gradually as James says, trying to improve the career prospects for our RSEs and similar professional career structures. Going from the old, "oh, you had a fixed term contract because you were funded for a specific research grant" to actually having a permanent post because we have enough work to support a team. To having a career structure where you’re not just stuck at one level of waiting for the next person up to leave and create a vacancy. But there is scope for career progression within the institution on a technical and research technology professional track, not just the traditional academic tracks.
[00:05:27] Jonathan Dursi: Terrific. And Chris?
[00:05:28] Chris Langridge: Hi. So yeah, I’m Chris Langridge department manager and as Donna alluded to I’m responsible for the day to day running of the department looking after the admin, the finance, the HR issues, and put into place the items discussed and making sure they work effectively within the department.
As James mentioned, ARC is a hybrid department and it sits as a professional services unit within UCL. But it has a research focus so this offers its own challenges and opportunities. So it’s trying to make sure that what we want to do can fit in with the way UCL’s used to working and doing things.
[00:06:07] Jonathan Dursi: Fantastic. James, could you tell me how hiring worked originally at UCL ARC?
[00:06:14] James Hetherington: Yeah. So let me wind the story right back if that’s alright, Jonathan, to 10 years ago when we first started the research software group.
The conception that we needed to come up with a better way of creating roles for research software developers working with UCL research groups, and create a better way of looking after those people, goes right back to a UCL IT reorganization in 2011, 2012, where there was a general hope from the universities to use more of its IT investment on research and on education, and less on the generics of the IT that any enterprise needs. And in particular that led to the founding of one of the predecessor organizations to UCL ARC, a research IT services organization, which was still then part of the IT department, but specifically had a small team for Research Software Engineers originally sold to the university really on the basis of getting more out of the investment in HPC, because there was somebody to help optimize the codes to work on the UCL clusters and do that kind of thing. And I was at that time working as a post-doc doing fluid dynamics work inside the brain and took on the challenge of trying to create this software function inside the then research IT services.
And we particularly hit on the importance of trying to grow that group beyond the me plus two centrally funded permanent team members through costing Research Software Engineers onto research grants.
But I pushed hard for the permission to not have the additional grant-funded resources be short-term posts, but to be longer term, permanent hires on the same basis as the rest of the teams. And there wasn’t a distinction between the team, we just had budget coming in. And started as a basis, which, for the first post was just going to HR and finance colleagues and showing them that there was the money there from research grants, that there wasn’t hardly any risk in that first post.
And it cascaded from there to about 12 people when I left UCL from that job and went to the Alan Turing Institute. And while I was at the Alan Turing Institute Jonathan continued to scale it. I think when I came back to run the new department, it was already… high twenties, Jonathan? You’ll have the facts more accurately.
[00:08:53] Jonathan Cooper: 30, I think.
[00:08:55] James Hetherington: Yeah, and it’s now mid forties. So that process has scaled remarkably well of costing people in our grants. It’s near impossible to hire 30% of an FTE of a good postdoc. Because we’ve got a staff pool model, we can do that. And because we’re offering a role that’s a permanent role rather than the usual precarity of these roles within within the post-doc system, I think we can attract talent. Which makes it an attractive mechanisms for people to use.
Within research grants, tricks over the years included spotting chances where a postdoc has left six months before the end of the research grant, how can the PI use that money. We’ve got an RSE knocking around and they could do some work on the code. And that’s allowed us to gradually scale it. And that was really one of the key things, my group, and that group in UCL then led by Jonathan, and equivalent groups in Manchester, South Hampton, and obviously the very long running Edinburgh Parallel Computing Center up in Edinburgh, that’s really the nucleus of the RSE story in the UK, using that method to scale things.
Because of that success, and ending up having a part of - being really candid, a research IT organization whose original founding principle was running data storage services and HPC services to the university - having more than half of its staff being grant-funded Research Software Engineers and Data Scientists making significant contributions to research grants, making significant contributions to research literature, we were able to persuade the university to take this momentum and build it into the new department, ARC, which is not part of the IT services organization spun out of that into a new department with both academic and professional services missions.
We think that the fact that we are running the services means that when we do research in the best ways to use data and computers and software to do science, it’s not just going to be a poster at a computer science conference and not have any impact, we have rich deep insight into the pain points and problems with people using data from computers and software to do research because we run those those platforms or services. So there’s a real synergy between ARC led research, our collaborations and consulting activity and our HPC and data management platform solutions and all three worked really tightly together.
There’s also significant appetite around the university for similar models of permanent staff pools in the skills of digital research, beyond programming. Collaborative data scientists working to get insight in data, but critically to other professions, which I’m building up within ARC. The research infrastructure developers built out of the teams of HPC, sysadmins, and so on that have always run our HPC services, working with research groups on the dev ops challenges and systems deployment challenges. So rather than having a research workflow running on a box in the corner of postdoc rooms, how do you build that out and turn it into well run infrastructure is code targeting a current cloud service or whatever is appropriate for that activity.
So I want to grow an outward facing collaborative team of Research Infrastructure Developers that you know, scales the same way as the Research Software Engineers. And then also Data Stewards is the next vertical, which is our people who are not data scientists who do the stats and machine-learning to get insight out of data, but organize and structure the data.
So that means helping labs make data find-able, accessible, interoperable and reusable by collaborating with them on the organization and structuring of that data. And then finally the fifth profession within ACT that Donna’s head of profession for is Professional research Investment and Strategy Manager, PRISM’s the new acronym. We didn’t start that. Wonderful person called Anja Roeding in Exeter? if I recall correctly, who started that acronym, but the idea that we often noticed that lots of research groups are appointing people as part of their research grants, grant funded to lead and coordinate research activities. And these are often people from the scientific community with a significant background, but then it’s their skills in agile delivery, it’s their skills in program management and community organization, that’s the way they contribute to the research effort.
And we find that in just the same way that agile project management often has a close synergy with the IT community the PRISM capability us a natural fit for a fifth profession within that set of professional technologists working with and as part of the scholarly community. So those are the sort of five professions we’re building within ARC. And knowing that we want to do that and having, a good 40 odd research software engineers 10 also data scientists, about five or six data stewards now growing faster and about 20 research infrastructure developers and Donna’s already talked about the PRISM team she’s building, which is, between 10 and 20 now.
That is the engine that’s going to deliver this digital transformation and research for UCL. So then the challenge is ,right, we’ve established the permanency. How do we do the rest of the things we’re going to need to do to make this new kind of career in the university work? How do we build towards the future? How do we create a smooth flow from entry grades through to very senior grades? How do we look after those professions? And critically, how do we make it so that it’s not just my central team within ARC itself but for all the people working to these kinds of work around the university, some of whom most of whom are still serial post-docs. How do we discover that and grow that within the profession as well? That I guess will be the topic of the rest of the call.
[00:14:38] Jonathan Dursi: Great. Yeah, exactly. Scaling this up beyond your own group is one of the many ways this group is going to have impact. So maybe Jonathan, I can ask you what did hiring for these permanent pool positions look like before. And what were the problems that caused for you and for the candidates?
[00:14:56] Jonathan Cooper: Yeah. So we started off basically having to show that we had enough money guaranteed from funded projects to support two years of a salary. And then we could get someone that was permanent position, albeit with the caveat of being funded for two years in the first instance, subject to continued funding being available. So, it was permanent, but with a slight caveat on it in that the university didn’t promise to underwrite the future risk of funding in the same way that they do for core funded posts.
And then as the group grew and built up a track record of successfully getting in funding, successfully having more staff, being able to show particularly future looking portfolio, these are all the grants that have gone in and are waiting a decision, on the basis for historical track record of success, this is the money that it looks likely we’re going to get.
Eventually we were able to persuade university finance team that the risk was sufficiently low, that people could be just hired on a straightforward, permanent post, when we had appropriate metrics showing the future forecasting of income.
And that would let us hire more standard grade RSEs. And we also had a job description for a senior level RSE that would do more of that sort of project management and line management of the team. And that was in place actually, before I joined as person number eight.
We needed more structure and hierarchy within the team at that point already. But we were still at the stage where we had to decide for each new post, okay, is this a standard RSE we’re going for? Is this a senior RSE we’re going for? Advertise one position. And it had to be a separate request to finance every time, " We think we’ve got enough income, can you check the forecast? Make sure you agree. Yes. Okay". Great, we can hire . That process in itself takes a while.
Then by the time you actually get the advert out, have had it out for sufficiently long, had to be out for a month in those days, if you wanted to hire outside of the UK. You’re then shortlisting, interviewing. It could easily be three, four months after the point to which you think I’ve got enough money for one person to actually be interviewing people for the role. By which point we often say "Ooh, actually, we’ve got enough money for two people now. "
And you think, "Actually I don’t necessarily want another standard RSE at this point. I’d quite like a standard one and a senior one, because I need to increase the management capacity as well as I’m growing the group." Or you get a pool of candidates at interviews and you say "I’ve got two very brilliant people here. I didn’t want to lose them at this point."
[00:17:32] James Hetherington: Or balancing expertise areas. So an HPC person and a web person or a geology background person, et cetera.
[00:17:40] Jonathan Cooper: Yeah. So, both the lack of flexibility and the slowness of University processes certainly made life more difficult than it would ideally be. And three conversations with different people in the HR teams discussing, "Are there ways in which we can do better than this."
And it’s interesting how the kind of institutional answers, even within the HR teams and different departments is very variable across an institution. So you quite often find somebody in one department that will tell you something is impossible, even if they’re in the HR team. And then the HR team in another department say yeah, actually, we’ve been able to do this.
It depends who you talk to, often. The HR team we had in IT services at the time were actually really good at finding work arounds and things that were allowed within the policies that weren’t always necessarily obvious or advertised as being available. And sometimes there were things that were permitted with the policy, but just the IT solution that the recruitment process used didn’t make it easy.
So we managed to make various improvements to the process over time through having these conversations. One was that we could do blended recruitment rounds. So this was something where I said, "Wouldn’t it be nice if I could advertise as an RSE and then appoint at standard or senior?"
But this doesn’t seem to be possible. I’m always asked just for which grade is it. "Oh, actually, we have seen one where they advertised for a seven or eight, depending on experience. Ooh, could you find out how they did that?" And, in fact, yeah, this is okay. And the system can handle two grades.
The system at time couldn’t handle three. So you couldn’t go across a wider range of experience, but blended across two grades was allowed. So once we discovered that was possible, we talked through how that works in practice, what you need to do in shortlisting and interviewing just to figure out which grade is someone most suited for, which grade are you going to interview them at? Then most of our RSE recruitment since were these blended recruitments and say we had the flexibility when we looked at the applicants that we’d got and at the team at the point where we interviewed. We could decide what we wanted to do.
We also discovered that, actually, even if we advertise, "we’ve got one post available". And then we get to interview and say we’ve got two brilliant candidates. And we’ve got enough money, we think, for two people, we could then put in a request to find out, say, "Actually we think you’ve got enough money for two. Now, can you check and approve? "
And because we are hiring to the same job description with the same process, we could then hire multiple people out of a single round, even if initially we only advertised for one. So that let us be a bit more flexible in the forward planning. As soon as we knew we could hire one person, we could start a recruitment round.
And then at the point where we’re actually interviewing, we can look who have we got, how many can we afford now? And take on more people as appropriate to get the right balance of skills and to meet the demands that we get.
Because our problem is always not finding enough money. It’s finding enough people to do the work that people are asking us to do. We’ve never been short of work in all the time that the group has existed. And usually we have too much work. And if people come to me and say, "I’ve got a grant that ends in three months, can you do a bit of work on it?" The answer is usually "No", or "Yes, if we postpone something else, we’ll have a think." Currently we are full for the next nine months or so.
[00:21:05] James Hetherington: Which would be the envy of many private sector consultants , as I keep reminding the team, when we get a bit panicky about it.
[00:21:12] Jonathan Dursi: So it sounds like step one was simply establishing a track record of having the money when you said you were probably going to have the money. And then once that happened, that unlocked a little bit of flexibility from finance. And then as you were hiring, you were going back and forth with HR, finding ways that they could work with you so that you could do these blended recruitments.
Chris, I know that you have been handling a lot of the back and forth with HR. I mentioned before this call that I have talked to many teams who have literally never spoken with their HR department except through filling out forms.
Can you tell me a little bit about how developing that working relationship with HR worked? And can you tell me what sorts of problems, what sorts of words, what sorts of terms of art resonated with HR?
[00:22:10] Chris Langridge: Yeah, of course. Firstly yeah, HR loves lots of forms, doesn’t it? That’s bread and butter.
We had different HR business partners over the last couple of years. To explain how this works is that HR is devolved into three main categories, the business partner oversees the policy and implementation of what we do. So we go to them with the standard questions about, how do you process this? What does that look like? And they tell us if we’re on the right track or if we need specific advice on matters.
Then you have the HR services department, which is a ticketing approach where you put in, I want to do this and then you’re all set or it comes back to you for more information. And then I fill in that missing data.
With recruitment, the process is mostly driven by finance within UCL. If you have the approval, then you’re able to just go ahead and recruit. So we’ve recently moved to new system called Talent Link which is a bit more open. We put the request in for these posts we’ve unlocked. And then we generate the advert, just straight out. It speeds this process up, because we’re now able to get a job advert out, approved, and available within a week, two weeks. Whereas previously, as Jonathan said, it was this process of going to the finance team, getting permission, going back, revise, and then posting for the advert. HR services would review that. And then it would come back with any queries. And then it gets approved.
So we’ve managed to streamline what was previously a week to 10 days process. That was driven mostly by a need to change within the university as part of a larger change piece. Building up a good relationship with both elements, the business partner, and those on the service desk, is fundamental to try and get these things actioned.
I feel personally a service desk approach for HR matters doesn’t always work. You lose that kind of personality where the HR team understand the problems and understand what it means to those people within the department.
To try and cross this barrier, I spent a lot of time, a little bit old fashioned, on the telephone, just calling them up and explaining why this ticket is important to me, how we can rectify it, what we can do to push this through and promote it as quickly as possible. This means that you’d get a personal buy in from the support on the ticket.
And you’re able to explain, actually I need this to go faster. I wanted to get a contract, because the contract requests have to be formally signed off. A lot of it’s automated, but it has to be formally signed off by a person. And trying to get somebody who is in the commercial sector, expecting a contract within a couple of hours, then hears that a contract can take a couple of weeks is quite unnerving. And you spend a lot of time trying to hire the correct person with a correct skill set. And the last thing you want is for your ticket to be stuck in the queue. It’s like issues with delays in starting.
There have also been, as Jonathan points out, there’s been a lot of issues around visas. And that’s been another thorn in my side. We’re seeking advice on how we, as a department, can collate as much information, and what point, to ensure that this doesn’t delay the contract. So applicants have to do through the right to work check. UCL than has to offer a certificate of sponsorship to say that we want to employ this person, they meet all the requirements. We then send the candidate to a standard processing team. They checked through, make sure that everything matches up. And then we offer them a contract for which they can apply for a visa. So it can be prolonged and it’s a very niche market. So having personal contacts within each team, and with the with the visa team directly, really helps speed along these processes. It’s definitely got quicker over the last couple months, though.
[00:26:19] Jonathan Dursi: Yeah. As teams of technologists. I think we underestimate sometimes the power of just picking up that phone and having a call and, routinely speaking with someone as a person. It’s extraordinarily effective, certainly a lot faster than bouncing back and forth on a ticket.
[00:26:38] Chris Langridge: One of my challenges, Jonathan, has been doing all of this new department building in the post COVID working environment with sickness, lower levels of face-to-face contact on campus, then we would have enjoyed before the pandemic. Which hasn’t helped, I would say.
[00:26:58] Jonathan Dursi: So Donna, I know that you have been instrumental in setting up the promotion pipelines and the career paths. It sounds like the team started with an RSE and a senior RSE role and not much more than that. How did promotions go from RSE to senior RSE when someone was in the team? And it sounds like a lot of the job descriptions up to this point had been bespoke. What sort of problems did that cause? How did the work start towards having a little bit more of a robust career ladder?
[00:27:31] Donna Swann: Yeah. Like you say, we did have a multitude of job descriptions across the department. So when we formally became ARC we had seen that most of the staff had come from the central research RIT, as we were called. And they came over on whatever job description they accumulated or been given X years ago. So we started off an exercise across the department, quite a good consultation exercise, actually, whereby we use the Wiki and we basically started off and said, "Let’s have a look at the professions. Let’s look at the four the PRISM one is in flight still. So, it was the research software engineer, the research data scientist, the research infrastructure developer and data steward, that’s right.
So we have the four of them. We basically just started off and started to look at what was generic across each of those professions? We started to look at things by grade and we built up actually some really nice documentation across the Wiki. And we got everybody across ARC to start to contribute to that. So what does it look like to be a grade eight at a data steward? What does it look like to be a grade eight as a research infrastructure developer?
So we started off actually at the six, which we felt generic across all the professions. And then we started to say, okay, what does a seven look like, generically? But then also what does the seven look like for that profession? And then we built on that for the eight and we built up that on to the nine.
So then we’ve got this really nice kind of set of activities, it was quite clear, about how you got from that grade seven and what does it look like to be an eight? However, a lot of staff then starts to say what would those activities look like because of that generic way that we’ve done it.
So we then built another Wiki page and all members of ARC started to document. What do those activities look like in that leadership role at a nine or in that technical role as an eight? And that’s how we built it. So we did that over about a three month period. And then at some point we said, okay, that’s it, kind of end of consultation. This is what we think we’ve got. And then yeah, so it ended up with 16 JDs is that right? across the professions. And we had HR lined up ready to grade those for us.
Whereas it previously in the past, when you create a job and then you get it graded. You know what it’s like, back and forth, things take a long time. We actually set some sort of time with with HR to do that piece relatively quickly. We did have a couple of issues, but relatively quickly.
And then we have these 16 JDS and then we were able to say to all staff, okay. Over the coming weeks, start to talk to your line manager about which is the profession for you because it wasn’t clear. Some of them were quite clear. I’m a RID, and that’s what I’m going to be.
But we did say that, you might find your primary profession is X, but that you’re actually also doing part of Y so over the coming months, then those discussions were had with line managers and pretty much think everybody now for existing staff are now in a kind of profession with the right JD and obviously now when we go to recruit, it’s so much easier, we pick one of those. We don’t deviate. We use the advert as the focus.
So if we want somebody who’s got X experience or Y then we really hone in on that in the advert text so that we can keep that generic JD set and those professions. So, yeah, so that was that. And so then as part of that, we had to have that career framework in place so that we could do the promotions piece. And so last April time, May time, we started to plan what that promotions looks like.
So ISD, the central IT enterprise division, they were also running a promotions process. We worked quite well with those, with that department as well. We use the same framework, but then we looked at it from an ARC perspective, what we wanted from that promotions piece.
So there were some differences, things like they decided to do like a manager led promotions process. We did a sort of self referral process for promotions. There was slightly different criteria, and a different way that we mark our promotions, but yeah, we opened up the promotions for a couple of months, we got people to apply. We then held a panel in around, I think, August of last year. We had 13 applications across the department, and we sat down and we rigorously went through them, each application in turn. We spoke about each of those. And then at the end, we came up with a kind of final listing of those that were promoted. Some were actually given an increment instead, and some weren’t lucky to get promoted this time. And we fed back to those individuals about how they can, again, looking at those JDS, what was missing, what parts weren’t quite there. And we went from there and we’ll run that again this year.
[00:31:54] Jonathan Dursi: That sort of normalization of existing staff, I think it’s one of the things that gives teams the heebie-jeebies about doing this, because there are a lot of managers dreading that conversation. I have to explain to Bob he’s not as senior as he thought he was. Was that a big deal in the end?
Maybe Jonathan, as one of the line managers. How did that go?
[00:32:21] Jonathan Cooper: There was certainly was some nervousness, particularly for those that had been on a fairly bespoke job description before. So if they’re Research Software Engineers, wasn’t really a big deal, because the new JDs were actually really similar to what most of the staff were on already. There were a few specialist roles looking at particular technologies, but even, so the generic ones encapsulated that.
One of the things that we said right at the start, which I think helped was that nobody would be changing grade as a result of the new jobs. Yeah. So if people looked at what was on the Wiki during the consultation phase said "I don’t see myself as a grade eight RID, I don’t think I meet these criteria." Then the message was, "The criteria is wrong. It’s not you don’t measure up, so we need to fix the criteria."
[00:33:07] James Hetherington: There’s a boundary condition that everybody was appropriately graded prior to the exercise. That may not have been perfectly true in all cases but a little bit of slight inflation around the edges there was worth it to achieve the assimilation onto the career structure.
[00:33:26] Jonathan Dursi: Absolutely. That makes perfect sense. And so, so now you have this clarity. Of career paths. I’m curious to hear from you, James. From the ARC-wide point of view, how has this clarity helped? From your point of view, directing the team, but also conversations with new recruits or experienced staff, does this clarity help?
[00:33:52] James Hetherington: So I think, first of all, one of the things that is really helpful is having a head of profession reporting to me for each of the five professions. That gives people a sense of who is the lead mentor for that activity, which is different from the person task managing them on research projects.
And that’s a really important division of labor for me, building leadership team, but it’s also important investing in people. As I’m making a new department on this hybrid academic professional services model. With a lot of change and a lot of change continuing, where the roadmap four building the perfect engine for helping people using computers and data and software to do science, is a long and complicated thread. And we need people to feel safe within that and we need people to feel when I’m asking them to do lots, asking them to do more, there is a track by which that will be reflected in compensation, and in seniority, and in title.
We have a hugely expanding range of services and platforms. And the fact that we’ve got a way for people to see, okay. If I’m stepping up, and without necessarily getting an upgrade now, I’m willing to take on being a strategic lead for this platform or an operational manager for that platform. Taking that citizen leadership, is one of the things that we’re able to make clear, will be to your professional advantage, as time goes forward.
Now I really believe in if you see challenges or problems within a workplace as a member of the workplace at whatever grade, being proactive about fixing it. I don’t want people saying, "James, this is broken. Can your leadership team, fix it, please?"
The same as I have with software, which is pull requests welcomed. We have a Wiki which describes the operating procedures and processes of the department. And if something is wrong in our business processes. It is open to everyone of every level of seniority within the department, not just to complain, but to help me fix it. What I want people to understand is that level of being a responsible citizen within the workplace is one of the key things that gets you promoted around here, right? As well as technical talent.
So that’s really important to me, getting our house in order around this. It’s also incredibly valuable for me in terms of being a good place to invest in, for the university senior leadership.
I also want the opportunities for staff within the central function to be better than the equivalent opportunities for staff within the post-doctoral precariat around the University. Not because of the old fashioned IT centralizing force, everybody must work within the center, nobody’s allowed to have any …. I’m not coming from that place. I’m coming from a place of trying to bring the benefits of what we believe having digital research professionals can do for the scientific mission to as much of the university as possible, if the most attractive roles within the university for being a data steward for being a research infrastructure developer, for being a PRISM, within ARC. My chances of selling that model to the person who is the local research IT individual within a particular site is more effective. That will in turn results in efficiencies for the university. But I’d rather win those efficiencies that way than through the the usual university approach to this, which is a central diktat. Because part of the reason so many people doing this wonderful work are in a postdoc precariat is because departments don’t want to let them be seen to be in a professional role because if they do they know they get snaffled off by the center at some point when there’s a cyclical change of wind. So hiding the work as a as a serial postdoc, rather than creating a professional role in the old apartment, looks like a smart move when you know that there’s a danger of centralization. So, creating a profession, which I hope we’re starting to see this with other teams around the university, spread and have this have professional structure, but not have to centralize all the roles under just my org chart. When I created this job, I want, as I said to Geraint, look. I need to ensure I’m not just going to be measured on the number of people working for ARC. We are measured on making UCL a brilliant place to use data and software and computers to do science. That’s the objective. And I think this is a key step towards that.
[00:39:04] Jonathan Dursi: Fantastic. And maybe finish off with Chris. The first thing that caught my attention about the work that UCL ARC has been doing with career paths was these always open positions. I gather that all the work that we’ve described so far has made it much easier to work with HR, to have these positions permanently open.
[00:39:25] Chris Langridge: Yeah. So all of this ties in, really nicely into making this what w e term an evergreen advert, to be constantly listed on the website. Allow people to apply. So we’re talking everything from James and Jonathan’s financial’s. The ability to open up posts quickly. The whole blended job families. This was 6, 7, 8, 9. That’s being able to say in the text because one thing is that UCL professional services never had a promotions process.
This is the first year it’s been run for. So that you can talk about, at UCL, how you can start at grade six and you can work your way up to the head of profession within this. If you you show the talent, the skills and the abilities. It’s quite an appealing advert. And see that up all year. It means that you’re not worried about "Oh I must apply today". It is a case of I can strengthen my CV. I’m interested in this. And just the other day, I had a candidate who said, " Oh, I’m really sorry. I can’t come to the interview, I’ve just taken up a fellowship". I said, that’s ok, lets me know. Please don’t worry. We’re constantly recruiting. So you may see that still up." And the reply was "That’s amazing. I’ll definitely be looking this time next year when my fellowship ends ."
So we’re keeping people that we are spotting now as potentially interviewable candidates and saying, if you’ve got something else on, this would just grow. And eventually when we get the department at the operational level that james has as this vision, there’ll be that natural turnover of staff where people move on to bigger, better things or elsewhere in the University etc. And we don’t want to have this struggle that Jonathan spoke about right at the start of the meeting was where we don’t want to be putting projects on hold because we can’t get the staff to come in. And then because, from my perspective, all of that causes this financial headache, because we’ve got to draw this money in. And if we haven’t got the staff to do it, We’re not bringing in the funds and then we’ve got… how do we allocate and reallocate staff versus money? It is good to have is to have this process. And this has been a lot of hard work to bring us here. And certainly having the buy-in from HR representatives that understand what we need and how we want to grow has been truly useful.
[00:41:44] Jonathan Cooper: One of the things I forgot to mention, is how we got here. Chris said that is the first time UCL has had a promotions process for professional services staff. Traditionally the model is that professional services is shaped as these are the posts that the university needs. And so you create a vacancy when you can justify having a new post and then people apply for it.
Which isn’t promoting people in place.
And having discussions. Why does this work for academics? Why doesn’t it work for professional services? One of the rationale is actually that as academics become more senior, they’re actually bringing in more value to the University. And so they pay for themselves, essentially because they bring in money.
So making the argument, we can do that sort of thing as well. And actually a change of leadership in IT made that possible to bring in for the wider group. And we rode on their coattails. A similar thing works when you’ve got these really skilled professionals as well.
Keeping them in the institution means they’re working better. And they’re actually bringing in value for the institution themselves. I think having that viewpoint has really motivated being able to run promotions processes. But one of the stepping stones we were able to do before we got to that point, because I didn’t want all my grade sevens hitting the top of their grade and then leaving. Running out of senior staff who are home grown. Blended recruitment meant that I had a promotions process by the back door effectively. Because people who are just RSEs could apply for any blended recruitment round that came up. Because it also offered the senior level. And if they then were successful at an open recruitment at a senior level. Yep. Great. And I’ve not lost the recruitment opportunity and then have to go and advertise again to refill their post. Because I can put them in a grade 8 position and still at the same time, refill the grade seven from people we’ve interviewed at that level.
So it made it much easier to promote people without actually promoting them because they were getting a new post. And the cost of paying someone at a great eight role and great seven is very small. So if ever anyone was successful internally, I pretty much automatically had the financial approval to say, yeah we can still recruit another as well. The marginal cost is small.
[00:44:02] Jonathan Dursi: That’s terrific. Thanks so much. There’s a million questions I could ask as follow up to this. Like now that you like this conversation with someone who apply for a year, how do you maintain a candidate pipeline? What’s next? But yeah. This is about the end of our time together. So thank you very much, all four of you for taking the time.
I think this is going to be really interesting broadly. Great talking to you.
[00:44:26] James Hetherington: Thank you, Jonathan.
[00:44:27] Jonathan Cooper: Pleasure talking to you as well.
[00:44:28] Donna Swann: Bye!
